How individuals became groups

You are not an individual in any simple sense. You are around thirty trillion cells that gave up reproducing on their own, each running on power plants that were once free-living bacteria, and the arrangement holds together well enough that you think of it as one thing. That transition has happened repeatedly, and it is the least understood part of evolution.

The pattern that keeps repeating

John Maynard Smith and Eors Szathmary catalogued the moments where biology changed what counts as an individual. In each one, entities that previously reproduced independently became parts of a larger entity that reproduces as a unit, and could no longer go it alone.

Why this should not happen

Here is what makes these transitions genuinely difficult, rather than merely impressive.

Selection acts on whatever reproduces. If a cell in your body could reproduce faster by ignoring the rest of you, selection at the level of cells would favour exactly that, and it would spread through your body within your lifetime. That is not a hypothetical: it is roughly what cancer is.

So every transition faces the same obstacle. Cooperation is vulnerable to whichever member stops cooperating and takes the benefit anyway, and that member leaves more copies, and the arrangement collapses. Getting from independent units to a stable collective means solving that, and the solution cannot be foresight, because nothing here has any.

How it gets solved anyway

Three mechanisms recur, and they work by removing the incentive rather than by enforcing good behaviour.

Shared fate. If units can only reproduce by the whole reproducing, defecting gains nothing. A mitochondrion cannot leave and start over, so its only route to more copies is the cell succeeding.

A single-cell bottleneck. Multicellular organisms mostly start life as one cell. This means every cell in a body is nearly genetically identical, so a cell that cheats its neighbours is cheating near-copies of itself and gains almost nothing in the currency that matters. It also resets any cheating that did arise, each generation.

Division of labour. Once units specialise, they stop being interchangeable competitors. A nerve cell and a muscle cell are not rivals for the same role, and neither can perform the other's function, which makes independence not merely unattractive but impossible.

Notice that all three work by making defection unprofitable rather than by policing it. That is the general shape of every stable cooperative arrangement in biology, and it is why cooperation does not require anything resembling intention.

Why this is the open problem

No simulation has convincingly produced a major transition that was not, in some form, built in. This is a known and slightly embarrassing gap in artificial life, and it matters because these transitions are plausibly what makes biological evolution open-ended: each one creates a new level for selection to act on, and a new space of things that can be built out of the level below.

A world that cannot produce them is a world where the space of possible organisms is fixed at the outset, however large it is, which is the subject of open-ended evolution.

Where this project stands

Directly and without hedging: nothing resembling a major transition has occurred here, and the architecture is not positioned to produce one.

Organisms reproduce individually, from a single parent, and there is no mechanism by which two organisms could come to reproduce as a unit. The genome has a fixed number of genes and the network a fixed shape, so there is no way for a collective to acquire a description of itself that it passes on as a collective. That is the thing a transition would require, and it is absent.

What does exist is a weaker and real relative. Organisms can move ground, and the changes outlast them, so one organism's actions affect organisms it never met, including its descendants. That is interaction across generations through the environment rather than through inheritance, and it is covered in niche construction. It is not a step toward individuality, and it is not presented as one.